From: Ed Sinjiashvili <edsin@...>
Date: 2001-08-14T15:01:44+09:00
Subject: [ruby-talk:19678] Re: () overloading

ts <decoux@moulon.inra.fr> writes:

> >>>>> "E" == Ed Sinjiashvili <edsin@swes.saren.ru> writes:
> 
> E> ts <decoux@moulon.inra.fr> writes:
> >> >>>>> "E" == Ed Sinjiashvili <edsin@swes.saren.ru> writes:
> >> 
> E> class Proc
> E> def () (*args)
> E> call(*args)
> E> end
> E> end
> >> 
> >> Proc#call use () which call Proc#(), which call Proc#call which use ()
> >> which call Proc#(), etc ... :-)))
> E> Proc#call is an instance of class Method - so it can not call Proc#() -
> E> unless you somehow override Method#() to use it. 
> 
>  I'll write it another way (just replace () with tt)
> 
>    class Proc
>       def tt *args
>          call tt *args
>       end
>    end
I got the point. If we have () operator then it's evaluated in scope
of Proc instance before going to call. Giving more thought to it, I
can't think of any reasonable () overloading except in Proc
class. And it's looks more like a syntactic sugar than a
feature. There's no real need to make class callable since we got
blocks which are first class functors. Thanx to everyone for helpful
insights. 

--
Ed Sinjiashvili